Denoising-Enhanced YOLO for Robust SAR Ship Detection

TL;DR

This paper introduces CPN-YOLO, an enhanced SAR ship detection framework that incorporates denoising, multi-scale feature extraction, and a novel loss function to improve accuracy and robustness in complex scenes.

Contribution

The paper presents a novel SAR ship detection method combining a learnable denoising module, PPA attention for multi-scale features, and Wasserstein-based loss for better generalization, outperforming existing detectors.

Findings

Achieves 97.0% precision and 98.9% mAP on SSDD dataset.

Outperforms YOLOv8 baseline and other detectors in complex scenes.

Demonstrates robustness against clutter and speckle noise.

Abstract

With the rapid advancement of deep learning, synthetic aperture radar (SAR) imagery has become a key modality for ship detection. However, robust performance remains challenging in complex scenes, where clutter and speckle noise can induce false alarms and small targets are easily missed. To address these issues, we propose CPN-YOLO, a high-precision ship detection framework built upon YOLOv8 with three targeted improvements. First, we introduce a learnable large-kernel denoising module for input pre-processing, producing cleaner representations and more discriminative features across diverse ship types. Second, we design a feature extraction enhancement strategy based on the PPA attention mechanism to strengthen multi-scale modeling and improve sensitivity to small ships. Third, we incorporate a Gaussian similarity loss derived from the normalized Wasserstein distance (NWD) to better measure similarity under complex bounding-box distributions and improve generalization. Extensive experiments on HRSID and SSDD demonstrate the effectiveness of our method. On SSDD, CPN-YOLO surpasses the YOLOv8 baseline, achieving 97.0% precision, 95.1% recall, and 98.9% mAP, and consistently outperforms other representative deep-learning detectors in overall performance.